EMG signals improved control of robotic leg prosthesis

Researchers found that the use of electromyographic signals in reinnervated nerves in the thigh muscles gave one transfemoral amputee better control of a motorized, robotic prosthesis, according to recent study results published in the New England Journal of Medicine.

In 2009, a 31- year-old man underwent a knee disarticulation amputation in which two nerve transfers were performed to prevent neuroma formation. The patient experienced discrete muscle contractions a few months after the targeted muscle reinnervation (TMR) surgery.

Researchers placed a cylindrical grid of 96 electrodes on the distal residual limb and collected electromyographic (EMG) signals while the patient attempted knee flexion, knee extension, ankle plantarflexion and ankle dorsiflexion of the missing limb.

The prosthetic socket employed a supracondylar suspension system. Researchers directed the amputee to walk with a robotic prosthesis with varying degrees of impedance for different ambulation modes, including walking on level ground, walking down a ramp with a 10° slope, and ascending or descending stairs with a reciprocal gait.

Thirteen mechanical sensors were located in the prosthesis. The patient completed more ambulation circuits during which transitions among ambulation modes were performed with the use of both EMG combined with mechanical sensor data to control the prosthesis, and with mechanical sensor data alone.

Study results showed that reinnervated hamstring muscles generated robust EMG signals, especially during contractions corresponding to ankle movements, and researchers noted marked coactivation of reinnervated muscles when the patient performed knee flexion.

Overall real-time error rate across all ambulation modes was 12.9% with the use of mechanical-sensor data only, which decreased to a real-time error rate of 1.8% when researchers added EMG information from reinnervated muscles. Researchers observed the patient ambulate robustly and transition easily with no critical perturbations among ambulation modes with the use of the TMR-enhanced system.

“The patient perceived that the TMR-enhanced system provided intuitive control during ambulation and non-weight-bearing activities. The reduced error rate enabled him to ambulate confidently and transition seamlessly among all modes with near-normal gait kinematics,” the researchers wrote.

Researchers noted several challenges to making the system clinically viable, including the need for the prosthesis to be more reliable, quieter and lighter to benefit more amputees.

For more information:

Hargrove LJ. N Engl J Med. 2013;369:1237-1242.

Disclosure: Levi Hargrove, PhD, reports being a consultant for SRA International. Todd A. Kuiken, MD, PhD, provided expert testimony for a law firm in a patient injury case and received payment for lectures from Firme Labs.

Leave a Reply

Your email address will not be published.